Correspondence to Dr Md Golam Abbas; [email protected]
STRENGTHS AND LIMITATIONS OF THIS STUDY
Retrospective design allows control of hospital stay duration.
Limited representation of the broader population due to single hospital-based nature and location specificity.
Short follow-up period preventing evaluation of long-term outcomes postdischarge for neonates.
Inability to assess outcomes of newborns taken against medical advice or referred to intensive care units, potentially leading to underestimated death rates.
Introduction
Globally, 5 million children under the age of 5 perished worldwide in 2020, with sub-Saharan Africa (2.7 million deaths, 54%) and South Asia (1.4 million deaths, 27%) accounting for the majority.1 Neonates comprised 47% of these deaths, up from 40% in 1990.1 Neonatal deaths accounted for nearly half of all deaths among children under the age of 5 in 2020, despite global neonatal mortality rates decreasing by 54% (from 37 neonatal deaths per 1000 live births in 1990 to 17 neonatal deaths per 1000 live births in 2020). This is because neonatal mortality rates declined more slowly than mortality rates for older children.1
Although neonatal deaths accounted for 62% of under-5 deaths in the region in 2020, with a neonatal mortality rate of roughly 23 deaths per 1000 live births, there has been a gradual improvement in newborn survival in South Asia.1 2 Bangladesh accomplished a 67% decrease in under-5 mortality rates under Millennium Development Goal 4 by 2015, helped by programmes to restrict the spread of diarrhoeal diseases, immunisation campaigns and the provision of vitamin A supplements.3–5 The government of Bangladesh and non-governmental organisations such as UNICEF, USAID and Save the Children established special care newborn units (SCANUs), as part of their commitment to enhancing neonatal survival in the nation. The goal of SCANUs is to prevent neonatal deaths resulting from avoidable causes such as infection and other birth complications.6 SCANUs had been set up in 74 healthcare facilities in Bangladesh as of 2022.7 However, with a neonatal mortality rate of 17 deaths per 1000 live births in 2020— considerably higher than the 2030 Sustainable Development Goal target of 12 deaths per 1000 live births—reducing neonatal mortality remains an important public health goal in Bangladesh.1 3
Birth complications associated with prematurity and intrapartum-related events are the leading causes of neonatal mortality globally.8 This holds for neonatal mortality in Bangladesh as well, where evidence shows that prematurity, perinatal asphyxia, sepsis and pneumonia are among the known causes.9 10 Studies conducted in community-based neonatal cohorts in Bangladesh have investigated factors such as the degree of prematurity and risk of neonatal infection to implement effective programmes to improve neonatal health.11 12 Although comparatively fewer studies have focused on characterising the same for SCANU cohorts in Bangladesh, previous research has suggested that neonatal mortality rates remained high even among neonates admitted to district hospitals and medical college hospitals (which are typically equipped with SCANUs and advanced care units), thus highlighting potential gaps between the medical needs of the neonatal population seeking care and the services available.13 Considering that SCANUs in Bangladesh have been established in multiple districts across the country with the goal of improving neonatal survival through the provision of life-saving treatments, characterising the outcomes of newborns and associated determinants in these units is crucial for measuring progress towards improved survival. In the present study, we aimed to describe the health outcomes of newborns hospitalised in a SCANU of a tertiary-level healthcare facility in the southern part of the country and to characterise potential demographic, diagnostic and clinical factors associated with their survival probability in the hospital. The findings of this study will be invaluable in focusing on areas of improvement and resource prioritisation to decrease neonatal mortality in low-resource countries around the world.
Methods
Study place, population and duration
A retrospective cohort study was conducted using health records of a SCANU situated in Patuakhali Medical College Hospital, Patuakhali, Bangladesh. The retrieved data covered a duration between August 2022 and December 2022. The SCANU of Patuakhali Medical College Hospital (a.k.a., Patuakhali Sadar Hospital) has been functioning as a secondary-level referral hospital for nearly a decade. It provides healthcare to 1 727 118 people living in 7 upazilas14 of the district, receives patients referred from upazila health complex (primary-level facilities) and refers patients when needed to tertiary level hospitals. Patients often come from the neighbouring districts, too. However, the SCANU provides services for babies born within (inborn) and outside (outborn) the hospital. The unit has facilities for resuscitation, oxygenation and exchange transfusion, incubators, radiant warmers and phototherapy devices. Oxygenation is maintained through oxygen cylinders and a central oxygen concentrator. Strict infection control measures are implemented by ensuring that each and every individual who enters the unit washes their hands with soap, nurses wash their hands before, and after touching neonates and use an aseptic technique for all procedures, septic patients are treated separately from aseptic patients and visitor entries are strictly controlled. Every neonate is initially checked by trained doctors and/or nurses of the unit for the presence of any danger signs. The danger signs looked for include—inability to suck, lethargy, convulsion, raised temperature, hypothermia, fast breathing, chest indrawing, yellow palms or soles and redness around the umbilicus. Services are made available 24 hours a day. However, it does not offer intensive care for children with critical health conditions. All ill neonates who are brought to the unit are examined and treated by trainee physicians under the supervision of specialist paediatricians. A record of the clinical findings, treatments and outcomes is kept for admitted newborns. Records of newborns who were referred to intensive care units or who left against medical advice were excluded from the study.
Information about newborns
Information collected from the SCANU registry for neonates includes data on age (early, mid, late and infants), sex (male or female), religion (Islam or Hinduism), birth week (preterm or term), birth weight (low, normal or high), place of delivery (inborn or out born) and delivery settings (healthcare or home/ambulance). Clinical and therapeutic characteristics such as capillary oxygen saturation (SpO2), the presence of danger signs, clinical diagnosis and the use of various medical interventions like antibiotics, radiant warmer, oxygen therapy and kangaroo mother care (KMC) were also documented. Danger signs were assessed by the duty doctors following a standard protocol followed in the centre. KMC involves positioning the neonate in the prone position in direct skin-to-skin contact with the mother, exclusive breastfeeding or tube feeding if unable to breastfeed and discharging the newborn when it becomes stable and gets at least 20–30 g of weight for three consecutive days. Antibiotics are empirically used for neonates with perinatal asphyxia, sepsis, low birth weight (LBW) and other diseases where a bacterial infection is suspected. Initially, ceftazidime and amikacin are given, switching to meropenem or ceftriaxone if there is no improvement within 72 hours. Oxygen is given to neonates with a SpO2 below 93% and is discontinued when saturation reaches 94% with spontaneous breathing.
Hospital outcome
We found four outcomes in the record: recovery and discharge, death, left against medical advice and referred. Recovery is defined when the newborn has no danger signs and is able to suck breast milk in an amount adequate to gain or maintain weight. If the children are taken away from the hospital by their parents or guardians against the physician’s advice, it is recorded as left against medical advice. Referral is usually done on approval from the head of the department to intensive care units or other hospitals due to the unavailability of required facilities in the hospital. A neonate is usually referred to when it requires ventilators to maintain oxygenation, has a birth defect, has hyperbilirubinaemia not amenable to phototherapy or has no visible improvement after giving all treatments for at least 3 days.
Data extraction and preparation
In addition to the paper-based documents (including individual patient’s admission files, treatment charts, investigation reports and the SCANU registrar), nurses maintain a Microsoft Excel file to record the data of admitted newborns. This latter process began as a part of the quality improvement initiatives in recent years. Initially, we retrieved information on 1079 newborns from the Excel file and checked the data for completeness and consistency. As there were no missing values or inconsistent reports, we focused on the recorded outcomes. As our study focused on in-hospital deaths of newborns, we excluded those who were referred and those who left against medical advice because no further information about their fate was available. Finally, a total of 930 newborn records were included for analysis. Five children aged between 29 and 50 days were admitted to the SCANU during this period and one died. Hence, we kept these records in the final analysis. Children’s duration of stay in the hospital was derived by subtracting the date of admission from the date of discharge or death. Dummy variables were created for causes, as many newborns had more than one cause for admission.
Data analysis
Data were cleaned and prepared for analysis in Microsoft Excel V.365. Data were analysed using Stata V.17 (StataCorp). The normality of quantitative variables was checked using histograms and the Shapiro-Wilk test. Bivariate analysis between two categorical variables was done using the χ2 test and Fisher’s exact test where appropriate. As the continuous variables (age, birth weight and duration of hospital stay) were not normally distributed, we used the Mann-Whitney U test to compare these variables across outcome categories. We used binary logistic regression analysis to determine factors associated with in-hospital deaths of newborns. Kaplan-Meier survival analysis was carried out to compare the survival probability of newborns across different characteristics. For multivariable Cox proportional hazard regression, covariates were chosen based on clinical relevance and findings of bivariate analysis and Kaplan-Meier survival analysis. A p<0.05 was considered significant for all statistical tests.
Results
Of the 930 newborns, 32 (3.44%) died in the hospital and 898 (96.56%) were discharged after recovery (table 1). The median age of the newborns was 1 day (IQR: 0.08–4), 739 (80.53%) were early neonates, 571 (61.40%) were male. 115 (12.37%) were born preterm (at gestational age 37 completed weeks) with 815 (87.63%) at term. 581 (62.54%) had appropriate weight for gestational age. 331 (35.63%) weighed <2500 g and 17 (1.83%) weighed >4000 g. The median duration of stay in the unit was 6 days. Most of the newborns (83.98%) were outborn and 79.6% were born in healthcare settings. A significantly higher proportion of preterm newborns died than term newborns (p=0.001). The birth weight of deceased newborns was significantly lower compared with those who were alive (p<0.001). Newborns born at home or ambulance faced death at a significantly higher proportion compared with those delivered in hospital (p=0.045).
Table 1Sociodemographic and birth-related characteristics of the neonates and infants by outcome (n=930)
| Characteristics | Total | Outcome | P value | |
| n (%) | Recovered | Died | ||
| n (%) | 898 (96.56) | 32 (3.44) | ||
| Age (days), median (IQR) | 1 (0.08–4) | 1 (0.08–5) | 0.12 (0.06–1) | 0.005 |
| Age groups, n (%) | ||||
| Early neonate (≤7 days) | 739 (80.53) | 708 (95.81) | 31 (4.19) | 0.005 |
| Mid neonate (8–14 days) | 71 (7.63) | 71 (100.00) | 0 (0.00) | |
| Late neonate (15–28 days) | 115 (12.37) | 115 (100.00) | 0 (0.00) | |
| Infant (>28 days) | 5 (0.54) | 4 (80.00) | 1 (20.0) | |
| Sex, n (%) | ||||
| Female | 359 (38.60) | 347 (96.66) | 12 (3.34) | 0.896 |
| Male | 571 (61.40) | 551 (96.50) | 20 (3.50) | |
| Religion, n (%) | ||||
| Islam | 887 (95.48) | 855 (96.39) | 32 (3.61) | 0.210 |
| Hinduism | 42 (4.52) | 42 (100.00) | 0 (0.00) | |
| Birth week, n (%) | ||||
| Preterm (<37) | 115 (12.37) | 105 (91.30) | 10 (8.70) | 0.001 |
| Term (≥ 37) | 815 (87.63) | 793 (97.30) | 22 (2.70) | |
| Birth weight (gram), median (IQR) | 2660 (2300–3000) | 2670 (2310–3010) | 2220 (1627.5–2700) | <0.001 |
| Birth weight groups (gram), n (%) | ||||
| Low birth weight (<2500) | 331 (35.63) | 310 (93.66) | 21 (6.34) | 0.003 |
| Normal birth weight (2500–4000) | 581 (62.54) | 570 (98.11) | 11 (1.89) | |
| High birth weight (>4000) | 17 (1.83) | 17 (100.00) | 0 (0.00) | |
| Place of delivery*, n (%) | ||||
| Inborn | 149 (16.02) | 146 (97.99) | 3 (2.01) | 0.297 |
| Outborn | 781 (83.98) | 752 (96.29) | 29 (3.71) | |
| Delivery settings*, n (%) | ||||
| At healthcare settings | 740 (79.66) | 719 (97.16) | 21 (2.84) | 0.045 |
| At home/ambulance | 189 (20.34) | 178 (94.18) | 11 (5.82) | |
P value was determined by χ2 test, Fisher’s exact test and Mann-Whitney U test where appropriate.
Significant p values (<0.05) were highlighted in bold.
*Inborn: Born in the place of the study; Outborn: Born outside the place of the study.
Table 2 enlists the clinical characteristics of the newborns. The median capillary SpO2 was 94% (IQR: 88%–97%). Of all, 8.76% had multiple danger signs with a significantly higher proportion of death (p=0.001) compared to those with who had single danger sign. The most common primary diagnoses during admission to the neonatal care unit were neonatal sepsis (35.59%), perinatal asphyxia (35.48%) and LBW (36.24%). Other reasons for admission included neonatal jaundice (12.9%), respiratory distress syndrome (RDS) (9.89%), hypoxic ischaemic encephalopathy (7.53%), intrauterine growth retardation (5.16%), small for gestational age (4.95%), respiratory tract infection (3.55%), being born to diabetic mother (3.01%), hypoglycaemia (2.47%), meconium ileus (2.04%) and birth defect (0.97%). Newborns with neonatal sepsis, LBW, perinatal asphyxia, RDS and hypoxic ischaemia encephalopathy had a significantly higher proportion of death (p<0.001, p=0.003, p<0.001, p=0.004 and p<0.001, respectively.)
Table 2Clinical characteristics of the neonates and infants by outcome (n=930)
| Characteristics | Total | Outcome | P value | |
| n (%) | Recovered | Died | ||
| Capillary oxygen saturation (SPO2), median (IQR) | 94 (88–97) | 94.5 (89 – 97) | 63 (36.5–89) | <0.001 |
| Any danger sign, n (%) | ||||
| Yes | 897 (97.93) | 865 (96.43) | 32 (3.57) | 0.402 |
| No | 19 (2.07) | 19 (100.00) | 0 (0.00) | |
| Danger signs, n (%) | ||||
| Multiple | 79 (8.76) | 70 (88.61) | 9 (11.39) | 0.001 |
| Single | 823 (91.24) | 800 (97.21) | 23 (2.79) | |
| Unable to suck breast milk | 187 (20.73) | 180 (96.26) | 7 (3.74) | |
| Fast breathing | 184 (20.40) | 173 (94.02) | 11 (5.98) | |
| Subnormal temperature | 181 (20.07) | 180 (99.45) | 1 (0.55) | |
| Lethargy | 100 (11.09) | 99 (99.00) | 1 (1.00) | |
| High temperature | 81 (8.98) | 81 (100.00) | 0 (0.00) | |
| Chest indrawing | 43 (4.77) | 42 (97.67) | 1 (2.33) | |
| Convulsion | 39 (4.32) | 37 (94.87) | 2 (5.13) | |
| Umbilical infection | 8 (0.89) | 8 (100.00) | 0 (0.00) | |
| Clinical diagnosis | ||||
| Neonatal sepsis, n (%) | 337 (35.59) | 336 (99.70) | 1 (0.30) | <0.001 |
| Low birth weight (<2500 g), n (%) | 331 (36.24) | 310 (93.66) | 21 (6.34) | 0.003 |
| Perinatal asphyxia, n (%) | 330 (35.48) | 308 (93.33) | 22 (6.67) | <0.001 |
| Neonatal jaundice, n (%) | 120 (12.90) | 118 (98.13) | 2 (1.67) | 0.253 |
| Respiratory distress syndrome, n (%) | 92 (9.89) | 84 (91.30) | 8 (8.70) | 0.004 |
| Hypoxic ischaemic encephalopathy, n (%) | 70 (7.53) | 62 (88.57) | 8 (11.43) | <0.001 |
| Intrauterine growth retardation, n (%) | 48 (5.16) | 46 (95.83) | 2 (4.17) | 0.679 |
| Small for gestational age, n (%) | 46 (4.95) | 44 (95.65) | 2 (4.35) | 0.669 |
| Respiratory tract infection, n (%) | 33 (3.55) | 33 (100.00) | 0 (0.00) | 0.270 |
| Infant of diabetic mother, n (%) | 28 (3.01) | 28 (100.00) | 0 (0.00) | 0.620 |
| Hypoglycaemia, n (%) | 23 (2.47) | 22 (95.65) | 1 (4.35) | 0.557 |
| Meconium ileus, n (%) | 19 (2.04) | 18 (94.74) | 1 (5.26) | 0.489 |
| Birth defect, n (%) | 9 (0.97) | 8 (88.89) | 1 (11.11) | 0.271 |
| Number of clinical diagnosis, n (%) | ||||
| Single | 456 (49.19) | 446 (97.81) | 10 (2.19) | 0.039 |
| Multiple | 471 (50.81) | 449 (95.33) | 22 (4.67) | |
| Duration of hospital stay (days), median (IQR) | 6 (4 – 8) | 6 (4 – 8) | 2 (1 – 4) | <0.001 |
P value was determined by χ2, Fisher’s exact test and Mann-Whitney U test where appropriate.
Significant p values (<0.05) were highlighted in bold.
The most common treatments required for the newborns were antibiotics (98.93%) and radiant warmers (87.53%) (table 3). Other required treatments in decreasing order of proportions were oxygen therapy (36.13%), essential newborn care (34.90%), tube feeding (17.01%), phototherapy (11.82%), KMC (6.32%), resuscitation (4.03%) and closed incubator (0.65%).
Table 3Treatments provided to neonates and infants by outcome (n=930)
| Characteristics | Total | Outcome | P value | |
| n (%) | Recovered | Died | ||
| Antibiotics | 915 (98.39) | 883 (96.50) | 32 (3.52) | 0.461 |
| Radiant warmer | 814 (87.53) | 782 (96.07) | 32 (3.93) | 0.025 |
| Oxygen therapy | 336 (36.13) | 309 (91.96) | 27 (8.04) | <0.001 |
| Essential newborn care* | 311 (34.90) | 297 (95.50) | 14 (4.50) | 0.125 |
| Tube feeding | 158 (17.01) | 144 (91.14) | 14 (8.86) | <0.001 |
| Phototherapy | 109 (11.82) | 108 (99.08) | 1 (0.92) | 0.163 |
| Kangaroo mother care | 54 (6.32) | 47 (87.04) | 7 (12.96) | <0.001 |
| Resuscitation | 37 (4.03) | 29 (78.38) | 8 (21.62) | <0.001 |
| Closed incubator | 6 (0.65) | 5 (83.33) | 1 (16.67) | 0.190 |
P value was determined by χ2 test, and Fisher’s exact test where appropriate.
Significant p values (<0.05) were highlighted in bold.
*Included delayed cord clamping, thorough drying, assessment of breathing, skin-to-skin contact and early initiation of breastfeeding.
Kaplan-Meier survival analysis showed that newborns with LBW (<2500 g) had lower probability of survival than those with a birth weight ≥2500 g (p=0.004). Preterm newborns (<37 weeks) had significantly lower survival probability compared with term (≥37 weeks) newborns (p=0.022). The presence of RDS was associated with a significantly lower survival probability (p=0.002) over the same period (figure 1).
Enlarge this image.Figure 1. Kaplan-Meier survival analysis showing comparison of survival curves across (A) delivery settings (at home/ambulance vs at hospital), (B) birth weight (>=2500 g vs <2500 g), (C) birth time (preterm (<37 weeks) vs term (>=37 weeks)) and (D) respiratory distress syndrome (present vs absent). (P values were determined by log-rank test).
Multivariable Cox proportional hazard regression analysis revealed that after adjustment for age, birth week, dangers signs, causes of admission (including perinatal asphyxia, neonatal sepsis, LBW and hypoxic ischaemic encephalopathy) and treatments provided (including KMC, essential newborn care, closed incubator, tube feeding, oxygen therapy and resuscitation), the newborn’s birthplace (p=0.021), SpO2 at admission (p<0.001) and the presence of respiratory distress at admission (p=0.032) came out as independent determinant of in-hospital mortality of newborns (table 4). Those who were borne at home/ambulance had 2.90 times higher hazard (95% CI: 1.17 to 7.17) of dying in the hospital compared with those who took birth in the hospital. One-unit higher SpO2 at admission was associated with a 3% lower chance of dying (adjusted HR: 0.97; 95% CI: 0.95 to 0.98). Newborns with RDS at admission had 3.39 times higher odds of dying (95% CI: 1.11 to 10.35) at the hospital than those without the problem.
Table 4Factors associated with risk of death among neonates and infants admitted to SCANU (n=930)
| Variables | Reference category | Crude HR (95% CI) | P value | Adjusted HR (95% CI) | P value |
| Age (early neonate) | Mid/late neonate and Infants | 8.16 (1.11 to 59.89) | 0.039 | 2.05 (0.25 to 16.64) | 0.503 |
| Birth week (preterm) | Term | 2.36 (1.10 to 5.06) | 0.027 | 0.96 (0.32 to 2.88) | 0.946 |
| Birthplace (at home/ambulance) | At hospital | 2.01 (0.96 to 4.20) | 0.063 | 2.90 (1.17 to 7.17) | 0.021 |
| Danger signs (multiple signs) | Single signs | 3.91 (1.76 to 8.69) | 0.001 | 2.25 (0.85 to 5.95) | 0.102 |
| SpO2 | 0.95 (0.94 to 0.96) | <0.001 | 0.97 (0.95 to 0.98) | <0.001 | |
| Respiratory distress syndrome (present) | Absent | 3.33 (1.49 to 7.47) | 0.003 | 3.39 (1.11 to 10.35) | 0.032 |
| Perinatal asphyxia (present) | Absent | 4.03 (0.190 to 8.50) | <0.001 | 2.08 (0.70 to 6.14) | 0.184 |
| Neonatal sepsis (present) | Absent | 0.06 (0.01 to 0.42) | 0.005 | 0.24 (0.03 to 2.17) | 0.206 |
| Low birth weight (present) | Absent | 2.82 (1.35 to 5.86) | 0.005 | 2.15 (0.82 to 5.63) | 0.120 |
| Hypoxic ischaemic encephalopathy (present) | Absent | 3.93 (1.75 to 8.81) | 0.001 | 1.62 (0.54 to 4.85) | 0.388 |
| Kangaroo mother care (no) | Yes | 0.33 (0.14 to 0.80) | 0.015 | 0.45 (0.11 to 1.78) | 0.256 |
| Essential newborn care (No) | Yes | 0.69 (0.33 to 1.43) | 0.320 | ||
| Closed incubator (no) | Yes | 0.19 (0.03 to 1.43) | 0.108 | 0.54 (0.04 to 5.96) | 0.615 |
| Tube feeding (no) | Yes | 0.36 (0.18 to 0.72) | 0.004 | 1.02 (0.37 to 2.72) | 0.976 |
| Oxygen therapy (no) | Yes | 0.12 (0.04 to 0.30) | <0.001 | 0.50 (0.17 to 1.46) | 0.207 |
| Resuscitation (no) | Yes | 0.15 (0.06 to 0.33) | <0.001 | 1.11 (0.33 to 3.69) | 0.864 |
Significant p values (<0.05) were highlighted in bold.
HR, Hazard Ratio; SCANU, special care newborn unit; SpO2, oxygen saturation.
Discussion
Our study highlights the mortality outcomes among neonates admitted to this SCANU, where 3.44% died in the unit and 96.56% were discharged healthy. There is a considerable variation in newborn deaths in different newborn care centres in Bangladesh.15 16 Khan et al 15 reported a high proportion of case fatality (10.83%) which, however, was much lower than that reported by Kumar et al (39%).16 Similar variations in mortality rate can be traced to other resource-constrained settings. For instance, mortality rates of 4.57%, 11.4%, 16.95% and 23.1% of neonates admitted to newborn care settings were reported in Libya,17 India18 and Ethiopia.19 20 A complex interaction of multiple factors might be responsible for the observed geographical variation in the mortality of newborns.21 However, preterm birth and birth-related complications are some of the crucial determinants of death in newborns21 which could be improved by increasing the quality of healthcare.
We noted a male predominance among the admitted newborns. This, although did not contribute to the mortality of the newborns, is an important point to consider since the sex ratio at birth is nearly equal in Bangladesh.22 A similar higher proportion of males was noted in other centres of the country.15 16 23 The high number of male newborns admitted to the SCANUs suggests that either female children are less prone to problems requiring admission in the hospital or they are less preferred when deciding to choose between home remedy and hospital care. Studies support the latter hypothesis, indicating that male sex was identified as a significant determinant of seeking healthcare from qualified providers more than two decades ago.24 A recent study suggests that the disparity still remains.25
In this study, the most frequent causes of newborn admission in the SCANU included neonatal sepsis, LBW, perinatal asphyxia, neonatal jaundice and RDS. Although there were differences in the frequency of occurrence of various causes, studies from other centres identified the same prevalent reasons behind the newborn admissions.16 18 Among all the causes of admission, RDS came out as an independent determinant of death among newborns, which is concordant with the findings of Cavallin et al 19 and Yismaw and Tarekegn26 in Ethiopia. Preterm neonates are prone to develop RDS due to inadequate production of surfactants.27 Reduced surfactant leads to atelectasis of the lung resulting in reduced gas exchange. The optimum management of neonatal RDS involves the use of prenatal corticosteroids to reduce incidence and severity. This is followed by the use of oxygenation, breathing assistance, surfactant therapy and general care for premature infants in the optimal management process. However, management often cannot be started timely because of a lack of antenatal care, delay in bringing the newborn to the hospital in case of home delivery or unavailability of the services in nearby health facilities. We found that one-unit increase in the partial pressure of oxygen was associated with a 3% lower chance of death in the hospital signifying the importance of managment.
A notable finding of our study was that after adjustment for age, birth week, danger signs, perinatal asphyxia, LBW, hypoxic ischaemic encephalopathy and KMC, birthplace came as a significant predictor of mortality in SCANU. Newborns delivered at home or in an ambulance on the way to hospital were statistically significantly more likely to die than those who were delivered at hospital. This implies that newborns delivered at home are more likely to suffer perinatal complications than those who get institutional delivery and endorses the findings of previous studies conducted in a similar context.28 29
Our analysis revealed in addition to resuscitation and oxygen therapy, KMC and tube feeding could be lifesavers for newborns admitted to special care units. KMC has been proven to improve survival between enrolment and 28 days in premature infants in multiple clinical trials30 and is a good low-cost option for reducing neonatal mortality in low-resource settings. KMC ensures skin-to-skin contact immediately after births and provides a means of thermal regulation of newborns. In Bangladesh, the practice of skin-to-skin contact after birth was practised in less than one-fifth of deliveries.31 This can be increased through KMC at facilities. Nasogastric tube feeding is practised in neonatal units in the case of premature newborns who are unable to suck an adequate amount of breast milk on their own.32 Mothers are encouraged to express their milk and feed through tubes until neonates start to suck on their own properly.
The strengths of the current study included its retrospective design, which allowed for control of the duration of stay in the hospital. The limitations include its lack of representation of a wider population throughout the country because of its single hospital-based nature and location specificity. Also, because of the short follow-up, long-term sequelae after discharge of the neonates could not be evaluated. An evaluation of the outcomes of newborns taken against the advice or referred to intensive care units was not possible, leaving the possibility of a trivial underestimation of the death rates.
To summarise, the present study found that preterm LBW, sepsis, perinatal asphyxia and RDS were the major causes of neonatal admissions at Patuakhali SCANU. Less than 5% of admitted newborns died in the hospital. Home delivery and respiratory distress were significant determinants of newborn deaths at SCANU. The fact that the study was done in one of the most resource-limited settings will help planners optimise resource allocation for better care. The findings would specifically be of interest to similar settings in resource-limited countries. Based on the study findings, we recommend that necessary steps should be taken to increase institutional delivery to reduce newborn deaths in low-resource settings.
The authors are grateful to the authority for giving permission to use the data and to nurses and doctors for aiding the retrieval of the data from the database.
Data availability statement
Data are available on reasonable request. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics statements
Patient consent for publication
Not applicable.
Ethics approval
This study was approved by the Ethical Review Board of Patuakhali Medical College (Ref: 2023/896). Permission was taken from the authority before data extraction. The unit takes informed consent from the parents of the children before initiating treatment at SCANU. We followed the latest edition of the ethical principles laid out in the Declaration of Helsinki.
Contributors MGA, AH, MASK and MMR conceived and designed the study. MMR, SSD, MW and MZA collected data and entered it for analysis. MASK analysed data. MASK, UT, TZ, MGA and AH interpreted results. MASK, UT, TZ, MGA and AH wrote and reviewed the first draft. All authors reviewed and approved the final draft. MGA acted as guarantor.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
1 UNICEF. Levels and trends in child mortality: report 2021. 2021. Available: https://data.unicef.org/resources/levels-and-trends-in-child-mortality/ [accessed 30 Oct 2023 ]
2 UNICEF. Neonatal mortality. 2021. Available: https://data.unicef.org/topic/child-survival/neonatal-mortality/
3 United Nations. Country profiles. In: Sustainable Development Report 2022. Cambridge University Press, 2022: 84–494. Available: https://doi.org/10.1017/9781009210058.015
4 Asian Development Bank. MDG 4: reduce child mortality. 2015. Available: https://www.adb.org/sites/default/files/publication/175162/ki2015-mdg4.pdf
5 Ministry of Health and Family Welfare Bangladesh, Partnership for maternal newborn and child health, World Health Organisation. Success factors for women’s and children’s health: Bangladesh. World Heal Organ; 2015. 17.
6 USAID. Managing sick newborns at special care newborn units in district hospitals in Bangladesh. 2018. Available: https://pdf.usaid.gov/pdf_docs/PA00TFB6.pdf
7 The Business Standard. 74 special care newborn units opened at hospitals in 50 districts. Available: https://www.tbsnews.net/bangladesh/health/74-special-care-newborn-units-opened-hospitals-50-districts-495758 [Accessed 30 Oct 2023 ].
8 Liu L, Oza S, Hogan D, et al. Global, regional, and national causes of child mortality in 2000-13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet 2015; 385: 430–40. doi:10.1016/S0140-6736(14)61698-6
9 Chowdhury HR, Thompson S, Ali M, et al. Causes of Neonatal Deaths in a Rural Subdistrict of Bangladesh: Implications for Intervention. J Health Popul Nutr 2010; 28: 375–82. doi:10.3329/jhpn.v28i4.6044
10 Khatun F, Rasheed S, Moran AC, et al. Causes of neonatal and maternal deaths in Dhaka slums: implications for service delivery. BMC Public Health 2012; 12: 84. doi:10.1186/1471-2458-12-84
11 Mitra DK, Mullany LC, Harrison M, et al. Incidence and risk factors of neonatal infections in a rural Bangladeshi population: a community-based prospective study. J Health Popul Nutr 2018; 37. doi:10.1186/s41043-018-0136-2
12 Shah R, Mullany LC, Darmstadt GL, et al. Incidence and risk factors of preterm birth in a rural Bangladeshi cohort. BMC Pediatr 2014; 14: 1–11. doi:10.1186/1471-2431-14-112
13 Halim A, Dewez JE, Biswas A, et al. When, Where, and Why Are Babies Dying? Neonatal Death Surveillance and Review in Bangladesh. PLoS ONE 2016; 11: e0159388. doi:10.1371/journal.pone.0159388
14 Bangladesh Bureau of Statistics. Population and housing census 2022. 2022.
15 Khan MKA, Al Mahmud SA, Islam SAN, et al. Disease Pattern and Outcome of Newborn Admitted in SCANU of Sheikh Sayera Khatun Medical College & 250 Bed General Hospital, Gopalganj, Bangladesh. The Insight 2021; 4: 54–61.
16 Kumar AA, Lee KH, Pervez AFM, et al. Factors Associated with Neonatal Survival in a Special Care Newborn Unit in a Tertiary Care Hospital in Bangladesh. Am J Trop Med Hyg 2023; 108: 844–50. doi:10.4269/ajtmh.22-0302
17 Rajab AM, Ghareba AM. Neonatal Mortality Rate in the Special Care Baby Unit (SCBU) at Gharian Teaching Hospital. J Med Sci Clin Res 2013; 1: 195–208.
18 Rahman M, Bezboruah G, Bhoktiari M. Morbidity and Mortality Profile of Neonates Admitted in Special Newborn Care Unit in a Tertiary Care Hospital: A Retrospective Study. J Clin Diagnostic Res 2023. doi:10.7860/JCDR/2023/62937.17617
19 Cavallin F, Bonasia T, Yimer DA, et al. Risk factors for mortality among neonates admitted to a special care unit in a low-resource setting. BMC Pregnancy Childbirth 2020; 20: 722. doi:10.1186/s12884-020-03429-2
20 Tekleab AM, Amaru GM, Tefera YA. Reasons for admission and neonatal outcome in the neonatal care unit of a tertiary care hospital in Addis Ababa: a prospective study. RRN 2016; 17. doi:10.2147/RRN.S95455
21 World Health Organization. Newborn mortality. 2022. Available: https://www.who.int/news-room/fact-sheets/detail/levels-and-trends-in-child-mortality-report-2021
22 The World Bank. Sex ratio at birth (male births per female births). Bangladesh, 2021. Available: https://data.worldbank.org/indicator/SP.POP.BRTH.MF?locations=BD [accessed 01 Nov 2023 ]
23 Nahar J, Zabeen B, Akhter S, et al. Neonatal Morbidity and Mortality Pattern in the Special Care Baby Unit of BIRDEM. Ibrahim Med Coll J 2007; 1: 1–4. doi:10.3329/imcj.v1i2.2896
24 Ahmed S, Sobhan F, Islam A, et al. Neonatal morbidity and care-seeking behaviour in rural Bangladesh. J Trop Pediatr 2001; 47: 98–105. doi:10.1093/tropej/47.2.98
25 Azad R, Billah SM, Bhui BR, et al. Mother’s care-seeking behavior for neonatal danger signs from qualified providers in rural Bangladesh: A generalized structural equation modeling and mediation analysis. Front Pediatr 2022; 10: 929157. doi:10.3389/fped.2022.929157
26 Yismaw AE, Tarekegn AA. Proportion and factors of death among preterm neonates admitted in University of Gondar comprehensive specialized hospital neonatal intensive care unit, Northwest Ethiopia. BMC Res Notes 2018; 11: 867. doi:10.1186/s13104-018-3970-9
27 Yadav S, Lee B, Kamity R. Neonatal respiratory distress syndrome. 2023. Available: http://www.ncbi.nlm.nih.gov/pubmed/15965895
28 Lee H-Y, Leslie HH, Oh J, et al. The association between institutional delivery and neonatal mortality based on the quality of maternal and newborn health system in India. Sci Rep 2022; 12: 6220. doi:10.1038/s41598-022-10214-y
29 Altman R, Sidney K, De Costa A, et al. Is Institutional Delivery Protective Against Neonatal Mortality Among Poor or Tribal Women? A Cohort Study From Gujarat, India. Matern Child Health J 2017; 21: 1065–72. doi:10.1007/s10995-016-2202-y
30 Zhu Z, Wang X, Chen W, et al. The efficacy of Kangaroo-Mother care to the clinical outcomes of LBW and premature infants in the first 28 days: A meta-analysis of randomized clinical trials. Front Pediatr 2023; 11: 1–8. doi:10.3389/fped.2023.1067183
31 Al Fidah MF, Efa SS. Skin-to-skin contact and early initiation of breast feeding in Bangladesh: a cross-sectional study using MICS6, Bangladesh (2019) data. BMJ Paediatr Open 2023; 7: e002163. doi:10.1136/bmjpo-2023-002163
32 McInnes RJ, Shepherd AJ, Cheyne H, et al. Infant feeding in the neonatal unit. Matern Child Nutr 2010; 6: 306–17. doi:10.1111/j.1740-8709.2009.00210.x
© 2024 Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group. http://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ . Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Alam, Mohammed Zahidul 2 ; Thamima, Ummay 3 ; Wahiduzzaman, Md 4 ; Zerin, Tahmina 5 ; Das, Siddhartha Sankar 4 ; Rahman, Md Mostafizur 4 1 National Institute of Preventive and Social Medicine, Dhaka, Bangladesh
2 Save the Children, Cox's Bazar, Bangladesh
3 International Centre for Diarrhoeal Disease Research Bangladesh, Dhaka, Bangladesh
4 Patuakhali Medical College, Patuakhali, Bangladesh
5 Refugee Crisis Foundation, Cox’s Bazar, Bangladesh